MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641071155 A) filed by Vellore Institute Of Technology on June 08, 2026, for System And Method For Adaptive Segmentation Of Handwritten Text Using Variable Kernels.
Inventors include Arpan Garai; Saquib Hussain; Ashish Kumar; and Avinabha Chanda.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: The present disclosure proposes a system (100) for adaptive pre-processing, segmentation, and binarization of handwritten Indic script documents, including but not limited to Devanagari script, to enable accurate detection, extraction, and processing of handwritten text regions in unconstrained handwritten documents. The system (100) comprises a computing device (102) having a processor (104) and a memory (106) for storing one or more processor-executable instructions. The processor (104) is configured to execute plurality of modules (108) for adaptive handwritten text segmentation. The plurality of modules (108) comprises an image acquisition module (110), a pre-processing module (112), a binarization module (114), a connected component analysis module (116), an adaptive kernel estimation module (118), a band- wise morphological processing module (120), a refinement module (122), a geometric filtering module (124), and a text-line detection module (126). The system (100) enables cross-script applicability for structurally similar Indic scripts, supporting scalable and versatile document analysis.
Disclaimer: Curated by HT Syndication.
